UC Riverside

Repeated presentation of speech syllables can change identification of ambiguous syllables, a perceptual aftereffect known as selective speech adaptation (e.g., Eimas & Corbit, 1973). Adaptation to auditory speech syllables can change identification of auditory speech and adaptation to visual (lipread) syllables can change identification of visual speech. Investigations of potential crossmodal influences suggest that selective speech adaptation depends on shared sensory information between adaptors and test-stimuli (Roberts & Summerfield, 1981; Saldaña & Rosenblum, 1994; Samuel & Lieblich, 2014). These findings have been cited as support for theories suggesting the speech process treats auditory and visual speech differently (e.g., Diehl, Kluender, & Parker, 1985; Massaro, 1987; Samuel & Lieblich, 2014). However, the lack of crossmodal influences in selective speech adaptation is inconsistent with a large literature across other speech domains suggesting crossmodal influences occur early in the speech process. These crossmodal findings from other speech domains are often cited as support for theories suggesting the speech process treats auditory and visual speech the same, based on modality-neutral information available in both (for a review, see Rosenblum, Dorsi, & Dias, 2016). This dissertation provides a rigorous investigation of crossmodal influences in auditory and visual speech adaptation. Chapter 1 investigated whether crossmodal adaptation can influence perception of visually-influenced audiovisual phonetic percepts. Chapter 2 investigated whether the subtle (non-significant) crossmodal adaptation measured previously (Roberts & Summerfield, 1981) is replicable and significant when provided sufficient statistical power (using larger sample sizes). Chapter 3 used bimodal adaptors to investigate whether crossmodal adaptation can augment changes in speech perception following within-modality adaptation (when adapted and tested in the same sensory modality). The results of the dissertation suggest that selective speech adaptation is sensitive to modality-neutral articulatory information. However, crossmodal adaptation is subtle, requiring larger samples of participants to reach statistical significance. These subtle crossmodal effects may be unobservable following concurrent within-modality adaptation, perhaps suggesting a ceiling effect when adapting and testing in the same sensory modality.